docker-compose.yml

services:
  postgres:
    image: pgvector/pgvector:0.8.1-pg17
    container_name: docgrid-postgres17
    environment:
      POSTGRES_DB: ${DB_NAME:-app}
      POSTGRES_USER: ${DB_USER:-app}
      POSTGRES_PASSWORD: ${DB_PASSWORD:-local_password}
    ports:
      - "${DB_PORT:-55432}:5432"
    volumes:
      - postgres17-data:/var/lib/postgresql/data
      - ./docker/postgres/init/001-enable-vector.sql:/docker-entrypoint-initdb.d/001-enable-vector.sql:ro
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U \"$${POSTGRES_USER}\" -d \"$${POSTGRES_DB}\""]
      interval: 10s
      timeout: 5s
      retries: 10
      start_period: 30s
    networks:
      - docgrid-local

  minio:
    image: minio/minio
    container_name: docgrid-minio
    environment:
      MINIO_ROOT_USER: minioadmin
      MINIO_ROOT_PASSWORD: minioadmin1234
    command: server /data --console-address ":9001"
    ports:
      - "9000:9000"
      - "9001:9001"
    volumes:
      - minio-data:/data
    healthcheck:
      test: ["CMD", "mc", "ready", "local"]
      interval: 10s
      timeout: 5s
      retries: 5
    networks:
      - docgrid-local

  embedding-server:
    build:
      context: ./backend/embedding-server
      dockerfile: Dockerfile
    container_name: docgrid-embedding
    restart: unless-stopped
    mem_limit: "${EMBEDDING_CONTAINER_MEMORY_LIMIT:-4g}"
    cpus: "${EMBEDDING_CONTAINER_CPUS:-4.0}"
    ports:
      - "8000:8000"
    environment:
      EMBEDDING_PROVIDER_MAX_CONCURRENCY: ${EMBEDDING_PROVIDER_MAX_CONCURRENCY:-1}
      EMBEDDING_PROVIDER_MAX_QUEUE_SIZE: ${EMBEDDING_PROVIDER_MAX_QUEUE_SIZE:-1}
      EMBEDDING_PROVIDER_QUEUE_WAIT_TIMEOUT_SECONDS: ${EMBEDDING_PROVIDER_QUEUE_WAIT_TIMEOUT_SECONDS:-15}
    volumes:
      - huggingface-cache:/root/.cache/huggingface
    healthcheck:
      test: ["CMD", "python3", "-c", "import urllib.request; urllib.request.urlopen('http://localhost:8000/health/ready')"]
      interval: 30s
      timeout: 10s
      retries: 5
      start_period: 900s
    networks:
      - docgrid-local

  prometheus:
    image: prom/prometheus:v3.5.5
    container_name: docgrid-prometheus
    profiles: ["monitoring"]
    restart: unless-stopped
    mem_limit: "${PROMETHEUS_MEMORY_LIMIT:-512m}"
    cpus: "${PROMETHEUS_CPU_LIMIT:-0.5}"
    command:
      - "--config.file=/etc/prometheus/prometheus.yml"
      - "--storage.tsdb.path=/prometheus"
      - "--storage.tsdb.retention.time=${PROMETHEUS_RETENTION:-7d}"
    ports:
      - "127.0.0.1:${PROMETHEUS_PORT:-9090}:9090"
    # Host에서 bootRun으로 실행한 Backend Management 포트를 Linux에서도 같은 이름으로 찾는다.
    extra_hosts:
      - "host.docker.internal:host-gateway"
    volumes:
      - ./monitoring/prometheus/prometheus.yml:/etc/prometheus/prometheus.yml:ro
      - ./monitoring/prometheus/rules:/etc/prometheus/rules:ro
      - ./monitoring/prometheus/targets:/etc/prometheus/targets:ro
      - prometheus-data:/prometheus
    # 감시 대상과 알림 경로가 장애여도 수집과 규칙 평가는 독립적으로 시작한다.
    networks:
      - docgrid-local

  alertmanager:
    image: quay.io/prometheus/alertmanager:v0.33.1
    container_name: docgrid-alertmanager
    profiles: ["monitoring"]
    restart: unless-stopped
    mem_limit: "${ALERTMANAGER_MEMORY_LIMIT:-128m}"
    cpus: "${ALERTMANAGER_CPU_LIMIT:-0.25}"
    command:
      - "--config.file=/etc/alertmanager/alertmanager.yml"
      - "--storage.path=/alertmanager"
      - "--data.retention=${ALERTMANAGER_RETENTION:-120h}"
      # 단일 로컬 인스턴스는 HA gossip 포트를 열 필요가 없다.
      - "--cluster.listen-address="
    ports:
      - "127.0.0.1:${ALERTMANAGER_PORT:-9093}:9093"
    volumes:
      - ${ALERTMANAGER_CONFIG_FILE:-./monitoring/alertmanager/alertmanager.yml}:/etc/alertmanager/alertmanager.yml:ro
      - ./monitoring/alertmanager/secrets:/etc/alertmanager/secrets:ro
      - alertmanager-data:/alertmanager
    healthcheck:
      test: ["CMD", "wget", "--spider", "--quiet", "http://localhost:9093/-/ready"]
      interval: 10s
      timeout: 5s
      retries: 5
      start_period: 10s
    networks:
      - docgrid-local

  grafana:
    image: grafana/grafana:13.2.3
    container_name: docgrid-grafana
    profiles: ["monitoring"]
    restart: unless-stopped
    mem_limit: "${GRAFANA_MEMORY_LIMIT:-384m}"
    cpus: "${GRAFANA_CPU_LIMIT:-0.5}"
    environment:
      # 1. Bundled Grafana는 Loopback 전용 읽기 화면이며 저장소에 관리자 Credential을 두지 않는다.
      GF_SECURITY_DISABLE_INITIAL_ADMIN_CREATION: "true"
      GF_AUTH_ANONYMOUS_ENABLED: "true"
      GF_AUTH_ANONYMOUS_ORG_ROLE: Viewer
      GF_AUTH_DISABLE_LOGIN_FORM: "true"
      GF_USERS_ALLOW_SIGN_UP: "false"
    ports:
      - "127.0.0.1:${GRAFANA_PORT:-3000}:3000"
    volumes:
      - ./monitoring/grafana/provisioning:/etc/grafana/provisioning:ro
      - ./monitoring/grafana/dashboards:/etc/grafana/dashboards:ro
      - grafana-data:/var/lib/grafana
    healthcheck:
      test: ["CMD", "wget", "--spider", "--quiet", "http://localhost:3000/api/health"]
      interval: 10s
      timeout: 5s
      retries: 10
      start_period: 20s
    # 2. Prometheus가 잠시 중단돼도 Grafana가 먼저 기동하고 연결 복구를 기다린다.
    networks:
      - docgrid-local

  valkey:
    image: valkey/valkey:9.1-alpine
    container_name: docgrid-valkey
    restart: unless-stopped
    ports:
      - "127.0.0.1:${REDIS_PORT:-6379}:6379"
    volumes:
      - valkey-data:/data
    healthcheck:
      test: ["CMD", "valkey-cli", "ping"]
      interval: 10s
      timeout: 5s
      retries: 5
    networks:
      - docgrid-local

networks:
  docgrid-local:

volumes:
  postgres17-data:
    name: docgrid_postgres17_data
  minio-data:
  huggingface-cache:
  valkey-data:
  prometheus-data:
  alertmanager-data:
  grafana-data: